scholarly journals Cartas de control para monitorear variables multinomiales

Respuestas ◽  
2014 ◽  
Vol 19 (2) ◽  
pp. 93-100
Author(s):  
Luz Marcela Restrepo-Tamayo ◽  
Juan Carlos Correa-Morales

Antecedentes: La carta de control como herramienta de monitoreo de la calidad de un producto, permite estudiar la estabilidad de los procesos en el tiempo, contrastando dos hipótesis, una que expresa que el proceso se encuentra en estado estable y otra que lo niega. Su utilización ha sido masiva para variables continuas más no para variables categóricas, motivo por el cual es imperante el diseño de tales herramientas para ese tipo de variables. Objetivo: Proponer dos (2) cartas de control para procesos con variables multinomiales basadas en el valor-p resultado de la prueba de homogeneidad de proporciones, empleando la transformación chicuadrado para variables uniformes y la aproximación Wilson - Hilferty para variables chi cuadrado. Métodos: El desempeño de las cartas propuestas es estimado vía simulación considerando un proceso en Fase II y considerando incrementos en la primera categoría de 2%, 4% y 6% en la etapa de control. Resultados: La carta de control multinomial usando aproximación Wilson- Hilferty para variables chi cuadrado, provenientes de la transformación del valor-p, presenta un desempeño deficiente comparado con las cartas de control usando valor-p y usando transformación chi cuadrado al valor-p, pues tienen menor habilidad para detectar cambios pequeños. Conclusión: Proponemos dos cartas de control para monitorear variables multinomiales y, una vez estudiadas vía simulación, con base en la Longitud de corrida promedio (ARL) y la probabilidad de rechazar la hipótesis nula de igualdad de proporciones, se recomienda el uso de la carta de control usando valor-p,o equivalentemente, de la carta de control usando transformación chi cuadrado del valor-p.Palabras clave: carta de control, distribución multinomial, prueba de homogeneidad, valor-p. Abstract  Background: The control as a tool for monitoring the quality of a product, allows to study the stability of processes over time, contrasting two hypothesis, which states that the process is in stable condition and the other denies it. Its use has been massive for continuous variables but not for categorical variables, why it is imperative to design such tools for such variables. Objective: To propose two (2) control charts for variables multinomial processes based on the p-value test result for homogeneity of proportions using the chi square test for uniform processing variables and approximation Wilson - Hilferty for variables chi square. Methods: The performance of proposed charts via simulation is estimated considering a Phase II process and considering the first category increments of 2%, 4% and 6% in the control stage. Results: The multinomial control chart using Wilson-Hilferty approximation for variables chi square, from the transformation of value-p, has poor performance compared to the control charts using p-value processing and using chi-square p-value, as they have less ability to detect small changes. Conclusion: We propose two control charts to monitor multinomial variables and once studied via simulation, based on the average run length (ARL) and the probability of rejecting the null hypothesis of equal proportions, we recommend the control chart using value-p, or equivalently, the control chart processing using chi square p-value.Keywords: control chart, multinomial distribución, homogeneity test, p-value.

Circulation ◽  
2020 ◽  
Vol 141 (Suppl_1) ◽  
Author(s):  
Gloria Kim ◽  
Arati A Gangadharan ◽  
Matthew A Corriere

Introduction: Some approaches to frailty screening use diagnostic or laboratory data that may be incomplete. Grip strength can identify weakness, a component of phenotype-based frailty assessment. We compared grip strength as a reductionist, phenotype-based approach to frailty screening with comorbidity and laboratory-based alternatives. Hypothesis: Grip strength and categorical weakness are correlated with the modified frailty index-5 (mFI-5) and lab values associated with frailty. Methods: Weakness based on grip, BMI, and gender was compared with mFI-5 comorbidities and lab values. Patients with at least 3/5 mFI-5 comorbidities were considered frail. Lab data collected within 6 months of grip measurement was assessed. Associations were evaluated using multivariable models and kappa. Methods: 2,597 patients had grip strength measured over 5 months. Mean age was 64.4±14.6, mean BMI was 29.5±6.9;46% were women, and 87% white. Prevalent comorbidities included hypertension (28%), CHF (22%), diabetes (29%), and COPD (26%); 9% were functionally dependent. 34% were weak, but only 13% were frail based on mFI-5. Hemoglobin, creatinine, and CRP differed significantly based on weakness ( Table ). Laboratory data were missing for 36%- 95% of patients. Multivariable models identified significant associations between weakness, hemoglobin, and all MFI-5 comorbidities. Categorical agreement between weakness and frailty was limited (kappa =0.09; 95% CL 0.0641-0.1232). Conclusion: Weakness based on grip strength provides a practical, inexpensive approach to risk assessment, especially when incomplete data excludes other approaches. Comorbidity-based assessment categorizes many weak patients as non-frail. Table. Demographic, laboratory values, and comorbidities by categorical weakness based on grip 20 th percentile. Mean values for continuous variables by weakness adjusted for gender and BMI, p-value for T-test; frequency and total percent for categorical variables, p-value represents chi-square test.


2018 ◽  
Vol 25 (12) ◽  
pp. 1887-1891
Author(s):  
Malik Jamil Ahmed ◽  
Muhammad Nasir ◽  
Aamir Furqan

Objectives: To investigate whether the addition of dexamethasone and chloropheniramine to oral ketamine premedication affects the incidence of postoperative vomiting. Study Design: Randomized control trail. Setting: Department of Anesthesia and Intensive Care Nishtar Hospital, Multan. Period: March 2016 to March 2017. Methodology: After obtaining ethical approval ethical and review board of hospital. Data was entered in a computer software SPSS version 23.1 and analyzed for possible variables. Continuous variables were presented as mean and standard deviation like age, weight, sedation time, anesthesia time, admission time and PACU time. Categorical variables were presented as gender, ASA statusand postoperative vomiting. Student test and chi square test was applied to see association of outcome variable. P value of 0.05 was taken as significant. Results: Overall, 100% (n=80) patients were included in this study, both genders. The study group was further divided into twoequal groups, 50% (n=40) in each, i.e. Group K (Ketamine) group and group KD (Ketamine-Dexamethasone). The main outcome variable of this study was postoperative vomiting. In this study, Postoperative vomiting observed in 35% (n=10) and 10% (n=4) patients, for group K and group KD respectively. The difference was statistically significant (p=0.007). Conclusion: Addition of dexamethasone and chloropheniramine with ketamine as premedication reduce the incidence of postoperative vomiting.


2021 ◽  
Vol 30 (03) ◽  
pp. 147-151
Author(s):  
Alia Ahmed ◽  
◽  
Usman Anwer Bhatti

OBJECTIVE: The objective of this study was to compare visuospatial and psychomotor skills of second year pre-clinical dental students with final year dental students using an exercise in dentinal pin placement. METHODOLOGY:A total of 120 BDS undergraduate students who had completed second or final year Operative dentistry rotation were included. While students from second and final year who had not consented to participate or had missed the practical demonstration or whose dentinal pins were misplaced after becoming loose from the tooth were excluded. Participating students placed the dentinal pins, following which Adobe Photoshop (version CC 2014) was used to analyze the photographs of the taken radiographs in two dimensions. Parameters assessed were pulpal perforations, periodontal perforations and pin angulation. Independent sample t-test was used to compare continuous variables while chi-square test was used for testing association for categorical variables. RESULTS: Final year students fared better in all categories of pin placement except periodontal perforation which was the same for both years. Statistically significant difference in the angulation for pin placement were observed between the two student groups in mesiodistal direction (p value =0.001) and in buccolingual direction (p value <.001). CONCLUSION: There is a significant difference in the psychomotor and visuospatial skill of second year pre-clinical when compared with the final year clinical undergraduate students. KEYWORDS: curriculum, dental, learning, operative, students.


Author(s):  
Sujata Pradhan ◽  
Padmalaya Thakur

Introduction: Ovulation is the key event in Intrauterine Insemination (IUI) cycles. Monitoring ovulation prior to insemination will help to alter insemination time to improve pregnancy rate. Aim: To compare pregnancy rates and live birth rates in presence and absence of ultrasonographic features of ovulation before insemination in IUI cycles. Materials and Methods: This was a retrospective cohort study conducted in a Institute of Medical Sciences and SUM Hospital, Bhubaneswar. Three hundred eighty eight IUI cycles performed in the period of January 2017 to December 2018 were analysed. On the day of IUI prior to insemination, transvaginal ultrasonography was done 36-38 hours after ovulation trigger to document ovulation. Presumptive signs of ovulation were documented in 201 cycles (Group A) and there was no feature suggestive of ovulation in 187 cycles (Group B). In all the cycles, single insemination was performed at 38-40 hours after ovulation trigger. Baseline characteristics were compared. Mann-Whitney U test was used to compare continuous variables. Chi-square test and Fisher’s-exact test were applied to find out the differences in the categorical variables as well as the pregnancy outcomes among the groups. Pregnancy rate and live birth rate were considered as the primary outcomes. Results: Pregnancy rate (17.9% vs 18.2%, p-value=0.945) and live birth rate (17.9% vs 16.0%, p-value=0.625) were similar irrespective of ovulation status documented in ultrasonography performed before insemination. Conclusion: IUI cycle outcomes are independent of the ovulation status documented before insemination.


2021 ◽  
Author(s):  
Amay Banker ◽  
Martin Gerdin Wärnberg ◽  
Anita Gadgil ◽  
Bhakti Sarang ◽  
Ramlal Prajapati ◽  
...  

Abstract Background The spleen is protected by the ribs anteriorly, the vertebral column posteriorly and the pelvis inferiorly. Fractures to this bony cage may indicate a high-grade splenic injury necessitating splenectomy. We aim to determine whether fractures to the bony cage protecting the abdomen are associated with splenectomy. Methods We performed a subgroup analysis of patients with splenic injury from a prospective trauma registry study named ‘Towards Improved Trauma Care Outcomes’ (TITCO) in India. Out of the 16047 patients enrolled in the TITCO study, 267 patients with splenic injury were included. Categorical variables were analyzed using the chi square test and logistic regression was used to assess the significance of continuous variables. A multivariate analysis was performed on the factors deemed clinically most significant. Results Patients with a higher grade of splenic injury were more likely to require splenectomy when adjusted for other variables (p value < 0.05). Patients with fractures to the vertebrae or pelvis had reduced odds of splenectomy on unadjusted analysis [Odds ratio 0.43 (0.19–0.94)], but this was not significant when adjusted for other variables. Conclusion In contradiction to our initial hypothesis, we found that fractures to the bony cage protecting the abdomen were not significantly associated with the splenectomy.


Blood ◽  
2012 ◽  
Vol 120 (21) ◽  
pp. 4653-4653
Author(s):  
Muath Dawod ◽  
Mohammad Alhyari ◽  
Philip Kuriakose

Abstract Abstract 4653 Background: Thrombotic thrombocytopenic purpura (TTP) is a rare but serious disease that has been reported more frequently in African-Americans (AA). While recent reports have showed an increased incidence of TTP amongs AA patients, there is no data with regards to the patterns of clinical presentation in this racial subgroup and if there is any difference as compared to other racial subgroups. Patients and methods: In this retrospective study we reviewed characterstics and clinical presentation of 91 patients with the diagnosis of TTP. Sixty four patients were AAs (70%). The second largest racial subgroup was Caucasians (24 patients, 26%). Demographics, lab values and symptoms were compared between AAs and Caucasians. Continuous variables were compared using two-sided two-sample t-tests. Alternatively, the Wilcoxon rank-sum test was used if the data were not normally distributed. Categorical variables were compared using Chi-square tests. Results: AA patients were older [median age of 47.6 years as compared to 40 years, P-value of 0.057] and more likely to be females [68.8%, (n=44/64) as compared to 45.8%, (n=11/24), P-value of 0.048]. As for laboratory evaluation, AA patients were more likely to have positive ANA [44% (n=22/64) as compared to 14.3% (n=2/24), P-value of 0.042] and higher ESR (median of 75.5 as compared to 40.7, P-value 0.03). Other laboratory evaluation including hemoglobin, platelet counts and creatinine were comparable. The most common patterns of clinical presentation in AA patients were in the form of constitutional symptoms (fatigue, malaise and generalized weakness) which were seen in 67% of patients (n=43), followed by neurological symptoms (mainly headache and confusion) in 53% of pateints (n=34). Gastrointestinal (GI) symptoms (nausea, vomiting or abdominal pain) were seen in 44% of patients (n=28). This pattern was comparable to what is seen in Caucasian patients. Skin symptoms were reported less frequently in AA as compared to Caucasians (17.2% vs. 41.7%, p = 0.016). Conclusion: While AA patients are more likely to be females, older and have positive ANA, the clinical presentation of TTP is similar among different racial subgroups (African Americans and Caucasians). Disclosures: No relevant conflicts of interest to declare.


Author(s):  
Hung-Chih Chen ◽  
Hung-Yu Lin ◽  
Michael Chia-Yen Chou ◽  
Yu-Hsun Wang ◽  
Pui-Ying Leong ◽  
...  

The purpose of this study is to evaluate the relationship between hydroxychloroquine (HCQ) and diabetic retinopathy (DR) via the national health insurance research database (NHIRD) of Taiwan. All patients with newly diagnosed type 2 diabetes (n = 47,353) in the NHIRD (2000–2012) were enrolled in the study. The case group consists of participants with diabetic ophthalmic complications; 1:1 matching by age (±1 year old), sex, and diagnosis year of diabetes was used to provide an index date for the control group that corresponded to the case group (n = 5550). Chi-square test for categorical variables and Student’s t-test for continuous variables were used. Conditional logistic regression was performed to estimate the adjusted odds ratio (aOR) of DR. The total number of HCQ user was 99 patients (1.8%) in the case group and 93 patients (1.7%) in the control group. Patients with hypertension (aOR = 1.21, 95% CI = 1.11–1.31) and hyperlipidemia (aOR = 1.65, 95% CI = 1.52–1.79) significantly increased the risk of diabetic ophthalmic complications (p < 0.001). Conversely, the use of HCQ and the presence of rheumatoid diseases did not show any significance in increased risk of DR. HCQ prescription can improve systemic glycemic profile, but it does not decrease the risk of diabetic ophthalmic complications.


2021 ◽  
Vol 42 (Supplement_1) ◽  
pp. S162-S163
Author(s):  
Jennifer B Radics-Johnson ◽  
Daniel W Chacon ◽  
Li Zhang

Abstract Introduction Burn camps provide a unique environment and activities for children that have experienced a burn-injury. Positive outcomes from attending burn camp include increased self-esteem, decreased feelings of isolation and a greater sense of self-confidence. In a 3-year retrospective review of camper evaluations from one of the largest and longest running week-long burn camps in the nation for ages 5–17, we aimed to assess if a child’s gender, age, TBSA or ethnicity affected the impact that burn camp had on a child. Methods A 3-year retrospective review of a Burn Camp’s camper evaluation forms was conducted for campers that attended burn camp between 2017–2019. Camp rosters were reviewed to determine the camper gender, age, TBSA and ethnicity. Camper self-evaluation forms completed at the end of each camp session were reviewed to record camper responses to questions regarding their opinions on the impact camp had on them as well as how camp will impact their lives once they return home. Categorical variables were summarized as frequency and percentage, and continuous variables were described as median and range. To check the relationship between two categorical variables, Chi-square test was used. To compare the continuous variable among groups, Kruskal-Wallis ANOVA was used. Statistical significance was declared based on a p value&lt; 0.5. Results Within 2017–2019, there were 413 camper records. Participants’ demographic characteristics are summarized in Table 1. There were 208 males (50.3%) and 205 females (49.6%). The median age of campers were 11.86, 12.44 and 12.45 for 2017–2019, with the range from 5.16 years to 17.96 years. The median TBSA were 20, 20 and 18 for 2017–2019, with the range from 0.08 to 90. Collectively there were 47.7% Hispanic (n= 197); 24.2% Whites (n=100); 13.1% Black (n= 54); 4.6% Asian (n=19) and 7.7% Other (n=32). There were 395 camper self-evaluation forms submitted. Results of three questions there we were interested in are summarized collectively in Table 2. 57% of campers responded, “Yes, Definitely” to the question “After going to this event, will you feel more comfortable being around your classmates or friends?” 54% responded, “ Yes, Definitely” to the question “Do you feel more confidents in sharing your burn story with others when returning home?” and 51% responded “Yes, Definitely” to “Did you learn anything that will help you when you return home?” Conclusions In analyzing the camper responses, there was no statistically significant difference in responses comparing gender, age, TBSA or ethnicity.


2021 ◽  
Vol 80 (Suppl 1) ◽  
pp. 734.1-734
Author(s):  
S. Maguire ◽  
F. B. O’shea

Background:Previous research in axial spondyloarthropathy(axSpA) has shown this population to have a high prevalence of depression. This co-morbidity has been previously shown to impact disease activity in patients with rheumatic disease.Objectives:The purpose of this study was to screen for early signs of depression using two validated tools, the Patient Health Questionaire-9 (PHQ-9) and the Hospital Anxiety and Depression Scale for depression (HADs-D) in patients with known axSpA.Methods:AxSpA patients attending the Rheumatology department in St James’ Hospital between February and October 2020 were invited to take a self-administered survey which included the PHQ-9 and the HADs-D. Scores from the HADs-D yielded a numerical result which was then categorised as normal, borderline or abnormal. PHQ-9 numerical results were categorised as normal, mild, moderate, moderate/severe or severe. Patients with a known diagnosis of depression were excluded. In addition to baseline demographics, patient reported outcomes from the clinic visit were also recorded.Data analysis was performed using IBM SPSS version 26. Continuous variables were recorded as means, categorical variables as frequencies with percentages. A one-way analysis of variance analysis (ANOVA) was used to determine significance of variation in outcomes between patient outcomes as determined by the HADs-D and PHQ-9. A p-value of <0.05 was deemed significant. Consent was obtained prior to participation. Approval was received from the St James’/Tallaght Hospital Joint Ethics Committee.Results:In total 71 axSpA patients took part in the survey. The population was 70.4%(50) males and 29.5%(21) female, with an average age 47.9 years and mean disease duration 19.7 years (mean outcomes: BASDAI 4.08, BASFI 3.62, BASMI 3.54, ASQoL 6.79). Overall, 7 (9.9%) participants recorded abnormal HADs-D scores, while 17 (23.9%) recorded moderate to severe PHQ-9 scores indicative of underlying depression. AxSpA females had higher mean HADs-D scores (7.5 vs 4.8, p=0.01) than males, with abnormal scores in 19%(4) of females and 6% (3) of males. No significant differences were found in PHQ-9 scores between genders.Analysis revealed significantly worse BASDAI (6.27 vs 3.42, p<0.01) and AQoL scores (12.57 vs 5.26, p<0.01) in axSpA patients with abnormal compared to normal HADs-D scores. No significant differences were noted in BASFI, BASMI or baseline demographics. A similar pattern was noted on analysis of PHQ-9 scores, with significantly worse BASDAI (7.9 vs 2.55, p<0.01), BASFI (8.05 vs 2.33, p<0.01) and ASQoL (19.5 vs 2.62, p<0.01) noted in those scoring as severe compared to normal. No significant differences were detected in BASMI scores or baseline demographics.Conclusion:A high percentage of axSpA patients recorded high HADs-D and PHQ-9 scores concerning for undiagnosed depression. These patients were noted to have significantly worse disease activity and quality of life as compared to patients with normal scores. Clinicians treating axSpA should consider screening for depression in this population.Disclosure of Interests:Sinead Maguire Speakers bureau: Speaker fee from Jassen, Grant/research support from: Recipient of the Gilead Inflammation Fellowship Grant, Finbar Barry O’Shea: None declared


2020 ◽  
Vol 7 (Supplement_1) ◽  
pp. S244-S244
Author(s):  
Bharath Pendyala ◽  
Prasanth Lingamaneni ◽  
Patricia DeMarais ◽  
Lakshmi Warrior ◽  
Gregory Huhn

Abstract Background Neurocysticercosis is a Neglected Tropical Disease and an important public health issue. Our goal was to collect and analyze data regarding clinically significant gender differences among our Neurocysticercosis patients. Methods A retrospective chart search with ICD 9/ ICD 10 diagnostic code for Neurocysticercosis and neuroimaging suggestive of Neurocysticercosis was performed for clinical encounters in the hospital or affiliated clinics between years 2013–2018. After a careful chart review, patients who were clinically diagnosed with Neurocysticercosis were included in the study. T-test was used to compare means of continuous variables and chi-square test to compare proportions of categorical variables. Results Among 90 total patients included, male (49.4%) and female (50.6%) distribution were nearly identical. The mean age in females was found to be higher than males (52.5 vs 42.0, P &lt; 0.0001). Almost an equal number of males and females presented with either seizures (63.6% vs 57.8%, P= 0.85), headaches (25.0% vs 28.9%, p= 0.85), or other symptoms (11.4% vs 13.3%, p= 0.85). Males had more generalized seizures compared to females (60% vs 38%, P= 0.37), although this result was not statistically significant. Females were more likely to present with &gt; 1 lesion (82.2% vs 56.8%, P= 0.01). Males were more likely to have cystic lesions (64.7% vs 27.9%, P &lt; 0.001) compared to females who had more calcified lesions on presentation (65.1% vs 20.6%, P &lt; 0.001). Male patients were more likely to have contrast enhancement or edema surrounding the lesions (61.4% vs 33.3%, P= 0.01) and were more likely to require treatment with Albendazole/Praziquantel (75.8% vs 31.7%, P &lt; 0.001). Conclusion Although previously reported data is limited, there is a suggestion that there are gender differences in host immune response and that inflammation surrounding parenchymal lesions is more intense in females. This study suggests that men either present early in the disease phase or have different immune responses than women and require anti-parasitic therapy more frequently. More research in this aspect is needed. Disclosures All Authors: No reported disclosures


Sign in / Sign up

Export Citation Format

Share Document